The language we use online matters exactly as much as the language in books, magazines and newspapers. Because articles can be written and published much faster online, we tend to have more flexibility in the way we pick our words compared to traditional media. While a magazine can afford to have a dedicated editor, the average blogger can't rely on the fact that someone else will correct his/her mistakes. And the big online audience guarantees that errors will be noticed at scale. This is why although writing and publishing take one click, we have to be especially careful in our word design as part of our web design. Words take only a couple of bytes, but failing to appreciate that leads to lengthy content that fewer people will load, read and understand entirely. Just because words aren't as heavy as other media doesn't mean we should abuse them. Getting our point across with as few characters as possible is our responsibility if not a case of art.
A web designer can't rely on the fact that the client will send him/her an error-free content in the rush to replace some meaningless filler text. If bad content slips online, it's as much the designer's fault as it is the client's fault. Everything is a signature.
Language isn't static—it changes over time as a function of the current generations' beliefs and behaviors. It follows trends, but at the same time, to be universally accepted in the future, it needs to be trend-free. Trends constantly get replaced with new ones and when people can sense an old one that they no longer consider right, they'll simply walk away. By trying to sound more cool today, we make our message useless tomorrow.
Different words I discover every day prove that many people are ready to ruin the commonly accepted language if that will give them some kind of advantage. Here is a short list of examples: gamification, appification, pretail, freetail, moonshot, photoshopping, thingiverse, sketchplanation. Now, because I'm insane too, I decided to use some of them in an improvised sentence, so I came up with “Pretail appification moonshot the thingiverse.” (There might be even an unintentional dose of truth in the meaning behind it.) Although I know approximately what this means, it's not immediately obvious to everyone, thus not exactly a sign of good design. Here we can see the problem: the reason such words can exist individually is that we can derive their meaning from a crystal-clear context. They tend to be loaded with ambiguity to be more interesting and this is exactly what makes language hard to understand, especially with a missing context. What does photoshopping mean—is it buying photos or just working with Photoshop? How would we explain this to a kid when we don't know it ourselves? The main reason for us to have thousands of words in the dictionary is that each one has one main meaning, that is universally accepted and agreed upon. By using our own thought-out words, we practically say that we disrespect the universal language/culture, being ready to contribute to its decay merely out of self-interest.
If a programmer would write variables and functions with non-obvious names, the software would be less likely to work as expected where many people collaborate on a large project. Programming is hard not because of the technologies themselves, but because everyone feels extremely capable to develop their own syntax that will obsolete everyone else's work. We see it every day with libraries developed on top of other libraries that change the original philosophy in some way until they get themselves changed. As a result, web developers have to use a variety of libraries that seen on the surface use the same technology, but have internal conflicts in the languages of their implementation. Treating everything as a black box doesn't help either, because it would mean that we don't care about dissonances in our voice. A great designer always strives for consistency and refuses to put his name on a product that violates his convictions. He does not tolerate the sense of proud he sees in people going great lengths to make their project "take off" at the expense of breaking the base station for everyone else. We need to start speaking the same language again, just to stay relevant in the future.